The emergence on the Inner Asian steppe of regional confederacies of pastoral nomads has figured prominently in the early historical records of China and other Old World states. Current hypotheses differ as to whether such polities arose as the result of indigenous political processes or from the influence of sedentary neighbors. Models illustrating these hypotheses are often based on historical sources and are rarely designed for testing against the material evidence recovered by archaeology. The Baga Gazaryn Chuluu (BGC) survey is designed to test ideas for early steppe political development using regional survey data and excavation. The project is set in a marginal frontier area having characteristics suitable for the study of both internal and external economic and political processes to enhance our knowledge of these societies.

In order to test ideas concerning the emergence of nomadic socio-political complexity, survey and small-scale excavation data will be collected from the site of Baga Gazaryn Chuluu in the desert-steppe zone of the Middle Gobi. The site is characterized by an extrusive granite ridge rising 300 m above the surrounding plain and having surface, spring, and well-water resources concentrated in outlying foothills. BGC fulfills all criteria required for the project which include clear evidence of habitation sites (artifact scatters) and mortuary and ritual sites dating to the periods of interest, location in a marginal area on the edge of the steppe zone, and differentiation from surrounding areas by the presence of high-grade pastoral resources. Given the observed density of sites dating to the 1st millennium BC, we consider BGC to be an early central place appearing on the boundary of the steppe frontier.

BGC currently supports over 30 herding families who make annual movements of 10-15 km in orbit of the ridge system. Based on this range of movement, a 525 sq km full-coverage survey will be conducted in and around the ridge formation by two crews of 5 members each. The entire project will be carried out over a period of 2-3 seasons of which the 2002 project will be the first. The survey will include three levels of resolution: an intensive phase consisting of transects walked at 30-50m intervals (400 sq km), an extensive phase of 1 km blocks walked at 100m intervals plus judgmental examination (125 sq km), and reconnaissance of up to 50 km beyond the central survey area conducted by jeep. Jeep recon will be an important check for outlying, high-visibility sites representing secondary or boundary habitation areas. Ground visibility at BGC is 95-100% and based on daily survey totals from northern Mongolia, we expect to cover 130 pedestrian sq km during a 2 month season. The locations of archaeological sites of all periods will be recorded using GPS units and 1:12,000 topographical map enlargements. Additional site information will include a survey number, brief description, dimensions and orientation, local environment, artifact presence or absence, artifact types, and collection details. A small number of habitation site test excavations will be conducted using 1x1m screened units placed in areas having artifact concentrations. These excavations will test for stratigraphy, deposit depth, and sub-surface artifact densities to provide information for a mid-scale excavation strategy in future seasons. 1-2 burials will be selected and systematically excavated with soil screening to provide bronze items for provenience and status analyses, dateable ceramics for the refinement of chronology, and skeletal data for information on demographics and paleopathology. A geomorphological survey is also included in the research design to enhance knowledge of local landscape processes.
Evidence for the evaluation of hypotheses will be developed using 3 main analyses: spatial analysis of site data, provenience and style analysis of artifacts, and bio-anthropological and statistical study of mortuary data. A GIS database containing a regional digital elevation model and environmental information from multi-spectral satellite images will be used to test for evidence differentiating our models. Specific tests will determine (a) whether sites cluster around pastoral resources or along probable routes of movement and exchange, (b) whether outer boundaries are uniformly maintained or are less restricted to the north and south, demonstrating directional openness, (c) whether site size and count hierarchies for habitation, ritual, and mortuary sites are present and the nature of their distribution in relation to resources and boundary areas, and (d) the extent to which upper levels of these hierarchies spatially coincide suggesting stability in central places and disengagement from pastoral routes. Bronze artifacts recovered from both mortuary and habitation contexts will undergo laboratory analyses (ICP & TIMS) for chemical composition and lead isotope ratios to determine their regions of manufacture. Finally, burials will be analyzed according to grouping and location, size, the number and provenience of artifacts, and demographic data to understand nomadic systems of status differentiation.